Claim Missing Document
Check
Articles

Penerapan Metode Fuzzy Time Series Chen Orde Tinggi Pada Peramalan Nilai Tukar Petani Provinsi Gorontalo Muhammad, Nur Miftah; Hasan, Isran K.; Arsal, Armayani; Rahmi, Emli; Nashar, Laode
Jurnal Riset Mahasiswa Matematika Vol 5, No 4 (2026): Jurnal Riset Mahasiswa Matematika
Publisher : Universitas Islam Negeri Maulana Malik Ibrahim Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18860/jrmm.v5i4.41337

Abstract

Nilai Tukar Petani (NTP) merupakan salah satu indikator ekonomi yang digunakan untuk menggambarkan tingkat kesejahteraan petani dan kondisi sektor pertanian. Pergerakan nilai NTP yang bersifat fluktuatif memerlukan pendekatan peramalan yang mampu menangkap pola data secara memadai. Penelitian ini bertujuan menerapkan metode Fuzzy Time Series (FTS) Chen orde tinggi untuk meramalkan Nilai Tukar Petani di Provinsi Gorontalo serta mengidentifikasi model orde yang memberikan tingkat kesalahan peramalan yang paling tepat. Data yang digunakan berupa data bulanan NTP Provinsi Gorontalo periode Januari 2020 hingga Oktober 2025 yang terdiri dari 70 observasi dan diperoleh dari publikasi resmi Badan Pusat Statistik. Data dibagi menjadi 80% data latih dan 20% data uji menggunakan pendekatan pembagian berdasarkan waktu. Tahapan analisis meliputi penentuan himpunan semesta, pembentukan interval, proses fuzzifikasi, pembentukan Fuzzy Logical Relationship (FLR) dan Fuzzy Logical Relationship Group (FLRG), defuzzifikasi, serta evaluasi kinerja model menggunakan Mean Absolute Percentage Error (MAPE). Hasil analisis menunjukkan bahwa model FTS Chen orde dua menghasilkan nilai MAPE sebesar 3,3164% pada data uji, yang lebih kecil dibandingkan dengan model orde satu. Sementara itu, model orde tiga tidak dapat digunakan secara optimal karena tidak terbentuk hubungan fuzzy pada beberapa periode data pengujian. Hasil ini menunjukkan bahwa pendekatan FTS Chen orde dua dapat memberikan hasil peramalan yang relatif lebih baik pada data NTP yang dianalisis dalam penelitian ini.
Development of a Hybrid ARIMA–Fourier Series Model for Air Temperature Forecasting at the Gorontalo Climatology Station Suci Tilome; Isran K. Hasan; Agusyarif Rezka Nuha
Journal of Mathematics, Computations and Statistics Vol. 9 No. 1 (2026): Volume 09 Issue 01 (March 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcosv9i1.11462

Abstract

Air temperature is a key climatic variable that reflects environmental conditions and influences various human activities. Recent observations indicate a persistent upward trend associated with global warming, leading to greater variability in climate patterns. These changes highlight the importance of forecasting methods that can accurately represent the characteristics of air temperature time series to support planning and decision-making. Reliable prediction is therefore essential for understanding climate dynamics and anticipating potential environmental impacts. This study proposes an air temperature forecasting approach using a hybrid Autoregressive Integrated Moving Average (ARIMA) and Fourier Series Analysis (FSA) model. The ARIMA component is applied to model trend behavior and temporal dependence, while FSA captures the remaining seasonal patterns in the ARIMA residuals. By integrating these two approaches, the hybrid model aims to improve forecasting accuracy in the presence of both stochastic and periodic components. The results show that the hybrid ARIMA–FSA model achieves good forecasting performance, with a Mean Absolute Error (MAE) of 0.56, a Root Mean Square Error (RMSE) of 0.66, and a Mean Absolute Percentage Error (MAPE) of 2.07%. These findings indicate that the proposed model effectively represents air temperature dynamics and can be considered a reliable alternative for climate forecasting applications
Adaptive ANFIS–PSO Model for Forecasting Bird’s Eye Chili Prices in Gorontalo Province Nur Siyam Djibu; Isran K. Hasan; Agusyarif Rezka Nuha
Journal of Mathematics, Computations and Statistics Vol. 9 No. 1 (2026): Volume 09 Issue 01 (March 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcosv9i1.11473

Abstract

Bird’s eye chili is one of the strategic food commodities in Indonesia with high price volatility and a significant contribution to food inflation, particularly in Gorontalo Province. The dynamic and nonlinear characteristics of bird’s eye chili prices often hinder accurate forecasting when using conventional methods, thereby requiring an adaptive approach capable of capturing complex data patterns. Therefore, this study applies an Adaptive Neuro-Fuzzy Inference System (ANFIS) optimized using Adaptive Particle Swarm Optimization (PSO) to improve the accuracy of bird’s eye chili price forecasting. This study utilizes daily bird’s eye chili price data in Gorontalo Province from 1 January 2019 to 31 October 2025, obtained from the National Strategic Food Price Information Center (PIHPS). The ANFIS model is optimized using adaptive PSO to obtain optimal parameter values that address local convergence problems and parameter sensitivity commonly encountered in conventional ANFIS models. Model performance is evaluated using the Mean Absolute Percentage Error (MAPE). The results indicate that the adaptive ANFIS–PSO model achieves a MAPE value of 17.4487% on the training dataset, which decreases significantly to 5.0741% on the testing dataset. The testing MAPE value below 10% demonstrates that the proposed model has excellent generalization capability in capturing bird’s eye chili price fluctuations. These findings confirm that adaptive PSO-based parameter optimization effectively enhances ANFIS performance in modelling nonlinear and highly volatile time series data. The proposed forecasting model can serve as a reliable analytical tool to support decision-making and regional food price stabilzation policies in Gorontalo Province.